Review Article IeJSME 2014 8(1): 8-18

Decision making in hyperglycaemia seen in Kavitha Nagandla, Sivalingam Nalliah

Abstract: Delay in childbearing, family history of type Key words: , Pregnancy, Medical Nutrition 2 diabetes mellitus and obesity in childbearing years Therapy, increases a possibility of intolerance or overt diabetes in pregnancy which may remain unrecognised unless an oral glucose tolerance test is done. Diagnosis of hyperglycemic disorders in pregnancy The International Association of Diabetes and The fundamental weakness in categorisation of what Pregnancy Study Group (IADPSG, 2010) recommended is termed ‘gestational diabetes mellitus’ (GDM) lies in the detection and diagnosis of hyperglycaemic disorders women not being aware of prevailing diabetes when in pregnancy at two stages of pregnancy, the first stage they become pregnant (pre-existing diabetes mellitus looking for ‘overt diabetes’ in early pregnancy based (DM), both Type 1 (T1DM) and Type 2 (T2DM). on risk factors like age, past history of gestational The diabetogenic action is mounted by pregnancy diabetes and obesity and the second stage where hormones such as corticotrophin releasing hormone, ‘gestational diabetes’ at 24-28 weeks with 75 g oral and coupled with glucose tolerance test. Although the one step approach other factors inherently seen in pregnancy such as with 75 g of glucose offers operational convenience accumulation of fat. Weight gain out of proportion to in diagnosing gestational diabetes, there are concerns normal invariably contributes to increasing insulin raised by the National Institute of Health in the recent resistance with advancing gestational age even in the consensus statement, supporting the two step approach absence of T1DM and T2DM. When the conventional (50-g, 1-hour loading test screening 100-g, 3-hour oral definition of GDM of ‘abnormal glucose intolerance glucose tolerance test) as the recommended approach in pregnancy’ is applied it would clearly include a for detecting gestational diabetes. Medical nutrition proportion of patients who have undetected pre-existing therapy (MNT) with well-designed meal plan and DM. However, attempts at newer categorisations has not appropriate exercise achieves normoglycemia without met uniform consensus. The International Association inducing ketonemia and weight loss in most pregnant of Diabetes and Pregnancy Study Group (IADPSG, women with glucose intolerance. Rapidly acting insulin 2010) proposed a novel categorizstion of ‘overt and analogues, such as and aspart are safe in gestational’. Overt diabetes is when the fasting blood pregnancy and improve postprandial glycemic control in sugar >7 mmol/L and HbA1c >6.5 % or a random women with pre-gestational diabetes. The long acting blood sugar >11 mmol/L.1 These criteria would be analogues (Insulin detemir and glargine) though proven meaningfully applied when screening for GDM is done to be safe in pregnancy, do not confer added advantage in early booking especially in the first trimester based if normoglycemia is achieved with intermediate insulin on ‘risk criteria”.1,2 Delaying screening to conventionally (NPH). Current evidence indicates the safe use of recommended glucose testing using the 75 g glucose at glyburide and in the management of Type 24-28 weeks would effectively miss a proportion of GDM 2 diabetes and gestational diabetes as other options. who would benefit from therapeutic manipulation to However, it is prudent to communicate to the women maintain a euglycemic state avoiding the often dreaded that there is no data available on the long-term health complication of macrosomia, perinatal mortality and of the offspring and the safety of these oral hypoglycemic intrapartum complications. drugs are limited to the prenatal period. ‘Gestational diabetes’ is diagnosed when FBS IeJSME 2014 8(1): 8-18 >5.1 mmol/L but is less than 7.0 mmol/L at any gestational age. The traditional 75 g oral glucose load

Department of Obstetrics and Gynaecology, Clinical School, International Medical University, Jalan Rasah, 70300 Seremban, MALAYSIA

Address for Correspondence: Kavitha Nagandla, International Medical University, Jalan Rasah,70300 Seremban, MALAYSIA Email: [email protected]

8 Review Article – Kavitha Nagandla, Sivalingam Nalliah IeJSME 2014 8(1): 8-18

at 24-28 weeks should have at least one abnormal value mothers and their off springs with regards to childhood i.e. FBS <5.1,7.0 mmol/L, 1 hour >10mmol/L and 2 hour obesity and maternal metabolic complications. >8.5 mmol/L.1 The NIH panel identified priority research with At the moment, this one step approach with 75 g randomised trials that would evaluate the diagnostic glucose as proposed by IADPSG is recommended by threshold of adverse outcomes, comparing the outcomes WHO and has been adopted by countries outside the of the two diagnostic approaches (one step versus two United States. The American College of Obstetrician step) their cost effective analysis, patients’ preferences and Gynaecologist (ACOG) however still recommends and their psychological consequences of the diagnosis the two step approach (50-g, 1-hour loading test (medicalisation of pregnancy) that would change future screening and 100-g 3-hour oral glucose tolerance test) decisions on diagnostic techniques. as they are of the opinion that the one step approach will result in increase in healthcare costs with no evidence Pre- pregnancy care (PPC): A shared responsibility of clinically significant benefits on maternal and fetal 3 Women with T1DM and T2DM should have outcome. preconception planning to achieve and sustain glycemic As the debate continues, the National Institute of control that minimises the risks of fetal malformation Health proposed the NIH Consensus Development and spontaneous miscarriage. In a report from California Conference in March 2013 to reach a uniform opinion. Diabetes and Research project, it was stated that The NIH panel concluded that the two step approach congenital malformations are more frequent in T2DM would be the recommended diagnostic method for compared to T1DM due to unawareness of the risk and gestational diabetes due to concerns related to one lack of preconception care.8 This emphasises the need for step approach such as increased frequency in diagnosis the primary care providers to identify women with risk (from 5 to 15%), patient anxiety, increased clinic visits, factors for diabetes such as obesity, acanthosis nigricans, intensive monitoring, caesarean delivery and additional polycystic ovarian syndrome, hypertension, parents health costs. The conclusions were largely based on the and siblings with T2DM and promote contraceptive fact that there is still paucity of evidence that treatment measures so as achieve optimum glycaemic control.9 of mild gestational diabetes improves maternal and fetal In a systematic review and meta-analysis by Wahabi outcomes.4 By far the most important epidemiological et al (2012), the effectiveness and safety of pre-pregnancy study on pregnancy outcome was ‘The Hyperglycemia care in improving the rate of congenital malformations Adverse Perinatal Outcome (HAPO) Study’. and perinatal mortality in women with pre-gestational At the same time of the HAPO study, two randomised diabetes was evaluated. It was reported that controlled trials that used the same glucose threshold preconception care reduces the incidence of congenital values as in HAPO study compared active treatment malformations and lowers the glycosylated haemoglobin with the standard obstetric care for women with mild A1C (HbAIC) by 1.92%. This meta-analysis revealed GDM. Both the trials demonstrated improved outcomes that only 34-38 % of eligible women actually utilise the such as large for gestational age, or birth weight pre-pregnancy care services. The authors highlighted >90th percentile and preeclampsia in the treated group the importance of routine integration of PPC services in and the results are complementary to the HAPO study.5,6,7 reproductive age, prevention of unplanned Despite these encouraging results, the composite and the need for further research as priority to identify outcome of neonatal morbidity and caesarean delivery the methods that will encourage the diabetic mothers to has not consistently improved with treatment and there utilise the PPC services.10 is paucity of evidence of the long-term outcomes for

9 Review Article – Kavitha Nagandla, Sivalingam Nalliah IeJSME 2014 8(1): 8-18

Pregnancy Care provide sufficient calories through the day and need to be synchronised with either oral hypoglycemics and/or The ultimate goal of optimal care in the management insulin therapy. of diabetes (pregestational and GDM) is to achieve euglycemia throughout pregnancy and this is best The American Diabetic Association emphasises the attained through a multidisciplinary team approach need to provide adequate calories so as to promote optimal including diabetic physician, dietician, nurse educator weight gain and promote fetal growth. Women should and family members.11, 12 be discouraged to skip meals and avoid heavy meals so as to prevent sustained and elevated postprandial blood Medical Nutrition Therapy (MNT) glucose which appears to contribute to macrosomia. Medical nutrition therapy is a programmed therapeutic Normal pregnancies would require a minimum of 1800 strategy that should be adopted in combination with kcal/day. The role of a dietician is to advise on what exercise and pharmacological intervention.. Attention constitutes a balanced diet and how much calories are required, individualising the requirements based on to maternal weight gain is relevant in management 13, 14 through a well-designed meal plan and appropriate body type and activity. exercise. Although evidence is sparse by way of well- A guide on calories requirements is shown in Table 1, designed studies, it is prudent to focus on medical based on glucose goals, weight gain and nutrient intake. nutrition therapy as fundamental to the management of Table 1: Calorie needs of Gestational diabetes mellitus14 GDM. In both overt and gestational DM, MNT should be instituted as the sole strategy or in combination with Body Weight Kcal/kg/day pharmacological therapy. Clearly where DM is detected Ideal body weight 30 early or the patient has pre-existing DM, MNT alone may be insufficient for optimum glucose control. T1DM Overweight 22-25 and poorly controlled T2DM are categorised as being at Morbidly obese 12-14 (present body weight) higher risk of requiring pharmacological therapy from Underweight 40 the outset especially when they are also overweight. Islet cell plasticity promotes increased insulin production As it is vital to synchronise pharmacological in normal pregnancies especially when pregnancy therapy to MNT the meal plan should be well-spaced hormones contribute to increased insulin resistance in throughout the day with appropriate individual meal the second half of pregnancy. This physiological response calorie manipulation of the total calories needs. Again of islet cells of the pancreas is inadequate in overt DM it is imperative for patients to be educated to be aware and hence both MNT and pharmacological treatment that the carbohydrate content has a direct bearing on are mandated to achieve desirable blood glucose level. postprandial glucose best demonstrated by periodic assessment of capillary blood sugar using a self-monitored The meal plan should be such that adequate calories glucometer. Although in women with ideal body weight, are provided for sufficient energy through pregnancy adopting a formula of 50% carbohydrate with the rest based on gestational age, body constitution, occupation of the calories being derived equally from protein and and category of GDM. A typical meal plan would fat is feasible, some therapeutic manipulation would be constitute frequent well-spaced meals constituting two needed in those who are beyond the ideal body weight. main ones (lunch and early dinner) interspersed with A changed formula of complex carbohydrate of 40%, snacks at mid-morning, mid-afternoon, and supper prior protein 20% and 40% fat (with no more than 7 % coming to retiring at night. The goal is to ensure that meals from saturated fat) should be advised. The quality of

10 Review Article – Kavitha Nagandla, Sivalingam Nalliah IeJSME 2014 8(1): 8-18

food should be aligned to the likes of the patient and her insulin and insulin analogues preparations are available cultural background to improve adherence to meal plan. and the beef and pork insulin preparations were phased Although caloric restriction is not adopted in MNT out in 2003 and in 2005.17 in the normally built patient, the Dietary Reference Intakes advocate such a strategy in those who weigh Rationale of insulin therapy in pregnancy 14 beyond 30 kg/m2. Maternal glucose crosses the placenta by facilitated diffusion. Maternal insulin does not cross the placenta Pharmacological therapy for diabetes in pregnancy unless it is bound to IgG antibody. Menon et al (1990), Insulin therapy has been in vogue as to complement hypothesised that antibody bound insulin in the fetus MNT in the management of diabetes for over 30 years. is an important determinant of fetal outcome such as Since the discovery of insulin by Banting and Best in macrosomia that is independent of maternal glycemic 1922, rapid technological advances have led to the control.18 This led to the need for an exogenous insulin development of several types of pharmacological agents.15 therapy that is of low immunogenicity which would Apart from insulin, insulin analogues and certain groups minimise placental transport of insulin. However, of oral hypoglycaemics are being increasingly used in relevance of these insulin antibodies to neonatal medical practice. outcome is still unclear. Balsells et al (1997) in his study demonstrated that maternal insulin antibody levels Insulin therapy did not influence the fetal outcome.19 This was further supported by Mc Cance et al (2008) that the level of Human insulin and insulin analogues and human insulin specific antibodies were The earlier insulin preparations were derived from low throughout pregnancy and there was no correlation porcine and bovine sources that had variable and between birth weight and cord blood human insulin unpredictable efficacy due to impurities that were (p=0.1590). Insulin aspart antibodies were undetectable immunogenic. Adverse effects such as in the cord blood of all participants. This study provided unawareness are attributed to these insulin antibodies. new information on the use of insulin analogues in 20 The recombinant human insulin emerged in 1986 pregnancy. with the substitution of alanine for threonine on the porcine insulin sequence. However 80% of diabetics and evidence for efficacy and safety treated with subcutaneous insulin still demonstrated of insulin in pregnancy insulin antibodies that can significantly alter its To understand endogenous insulin physiology, the pharmacokinetics. The hypothetical explanation is concept of basal and bolus insulin need to be defined. that the insulin antibodies can serve as carrier and Basal insulin is background insulin that is released by prolong the effects of insulin or decrease its efficacy the beta cells of pancreas and is present throughout by neutralising its actions. It is identified that the the day and bolus dose is the insulin that is released in insulin binding sites of the antibodies vary among response to glucose from the meals (postprandial peaks). the individuals and therefore the pharmacodynamics. Simulating this mechanism of release of endogenous This resulted in exploration of exogenous agents that insulin is achieved by both basal (intermediate or long can closely mimic the physiological peripheral response acting insulin) and bolus dose (short or rapid acting of insulin by slightly modifying its amino acid sequence.16 insulin) either as multiple dose injections or continuous Thus insulin analogues were introduced with insulin subcutaneous pumps. Essentially, 40-50% of daily total lispro in 1996. Currently, only recombinant human dose of insulin replaced should be the basal insulin to

11 Review Article – Kavitha Nagandla, Sivalingam Nalliah IeJSME 2014 8(1): 8-18

cover for overnight and between meals and the other proline by aspartic acid in position 28. The glulisine has 50-60% is in the form of bolus dose of insulin so as to a double substitution that replaces aspargine by cover glucose rises due to carbohydrate that is consumed at position 3 and lysine by glutamic acid at position 29. and postprandial hyperglycemia.21 The effects of these modifications in the beta chains results in insulin analogues that form monomers which Types of insulin are rapidly absorbed upon administration and quick 24 There are currently four types of insulin preparations onset of action. depending on the time of onset and duration of action Regular human vs rapid acting insulin analogues: which include short acting insulin, rapid acting insulin evidence in pregnancy analogues, intermediate acting insulin and long acting insulin analogues (Table 2). For achieving near- The rapidly acting insulin analogues are comparable normoglycemia, these insulin preparations should to with respect to their immunogenic have similar pharmacokinetics as endogenous insulin. properties. The onset of action of rapid acting insulin This refers to their action which results in rapid rise in analogues is faster with earlier peak concentration and insulin after the administration, short duration of peak brief duration of action that prevents postprandial insulin levels, rapid decline of insulinemia for short hyperglycemia and late onset hypoglycaemia.24 However acting insulin and even insulinemia without peaks or only lispro and aspart have been investigated in pregnancy glycemic excursions in (blunting effect) intermediate and have demonstrated considerable safety profiles acting insulin.22 The optimum glycemic control is with no risk of teratogenicity.25 In a study by Jovanovic therefore achieved by combination of long acting and et al (1999) comparing the effects of lispro with regular rapid acting insulin as basal–bolus dosing mimicking insulin in gestational diabetes, lispro was undetected normal physiological insulin patterns. in the cord blood, indicating no placental transfer. The mean fasting postprandial levels and HbA1c were Regular human insulin the same in both groups with lispro demonstrating lower 26 Regular human insulin is short acting insulin that is hypoglycemic episodes. In a randomised control trial the least immunogenic. The specific problems related to in pre-gestational TIDM, insulin aspart was found to regular insulin is its slow onset of activity that results be superior to isophane insulin (NPH) in reducing the risk of postprandial glycemic excursions and delayed the in inadequate control of postprandial hyperglycemia, 27 long duration of action and potential for late postprandial postprandial hypoglycaemia. In an observation study on hypoglycaemia.23 safety of insulin lispro in TIDM, T2DM and gestational diabetes, which analysed 635 pregnancies over a period Rapid acting insulin analogues of 7 years, it was concluded that patient satisfaction was considerable with insulin lispro (P<0.05) with The rapid acting insulin analogues are insulin no difference in maternal or fetal outcomes, whether lispro, aspart and gluisine. The insulin analogues patients used regular insulin (n=138) or insulin lispro are produced by recombinant DNA technology. (n=75), and with lower antepartum HbA1c with insulin The lispro is prepared from E.coli and it differs from lispro (p<0.05).28 human insulin in the substitution of two amino acids in the beta chains. The amino acid lysine is substituted Intermediate acting Insulin at position 28 and proline at position 29. The aspart has the yeast Saccharomyces cerevisiae to replace the Neutral Protamine Hagedorn (NPH) still remains the basal insulin of choice during pregnancy due to

12 Review Article – Kavitha Nagandla, Sivalingam Nalliah IeJSME 2014 8(1): 8-18

its flexibility in adjusting the dosage in response to that glargine was associated with low incidence of calorie intake. Nocturnal hypoglycaemia is of concern macrosomia.30 The insulin and insulin like growth factor with NPH due to its peak action occurring 5-7 hours receptor- 1- binding properties, mitogenic properties of the dose that is less likely to reduce even with the of glargine were found to be increased in comparison consumption of bedtime snack. This is addressed by to other insulin analogues in a study by Kurtzhals et al shifting the pre-dinner NPH insulin to bed that delays (2000). This raises the concern of its safety in pregnancy.31 the peak action to early morning thereby avoiding the In a recent meta-analysis by Lepercq et al in 2012, overnight hypoglycaemia.29 exposure of long acting insulin analogue glargine, in all trimesters of pregnancies demonstrated no adverse Long acting insulin analogues maternal and fetal outcome compared with NPH The long acting insulin analogues are glargine and insulin. However, it was noted that the number of detemir. In glargine, glycine is substituted by aspargine women treated with glargine in the first trimester is small to provide conclusions on its safety with respect at position 21 and two arginine molecules are added at 32 the carboxyl end. In detemir, there is elimination of to teratogenicity. Regarding insulin detemir, a recent threonine and substitution by by acylation. randomised, controlled non-inferiority trial compared This process of amino acid modification shifts the pH to the efficacy and safety of insulin detemir versus NPH neutral and this delays absorption, self-aggregation and in pregnant women with TIDM. The primary outcome, therefore prolongs the duration of action.23 suggested non-inferiority of insulin detemir to NPH with respect to HBA1c at 36 gestational week.33 Neutral Protamine Hagedorn (NPH) insulin vs long Following this study results, detemir received Food acting insulin analogues: evidence in pregnancy and Drug Aadministration (FDA, USA) approval as category B drug from Category C. However, at the In a retrospective study by Callesen et al (2013), moment, with the available evidence, it is not prudent the glycemic control and pregnancy outcome was to initiate glargine in pregnancy unless more data is compared between insulin detemir and glargine in obtained from large randomised controlled trials and TIDM. It was concluded that both the long acting although the evidence for detemir is favourable, there is insulin analogues are safe in pregnancy with similar no compelling evidence for the routine use of this drug glycemic control and pregnancy outcomes except in pregnancy.34 Table 2: Types of insulin and insulin analogous and their pharmacokinetic properties24, 25 Name and Type of Insulin Onset of Action Peak Effect Duration of Action Short Acting Insulin 60 minutes 2-4 hours 6 hours Regular Insulin Rapid Acting Insulin Analogues (Bolus) Insulin lispro 15 minutes 60 minutes 2 hours Insulin aspart Intermediate Insulin 2 hours 4-6 hours 8 hours NPH (Basal) Long Acting Insulin Analogues (Basal) No peak effect for Insulin glargine 2 hours 12 hours Insulin detemir is 3-9 hours Insulin detemir Recommended Intervals of Dosing Regular Insulin: 60 minutes before each meal, Rapid Acting Analogues: At the start of each meal, Intermediate Insulin : Every 8 hours, Long Acting Analogues: Every 12 hours

13 Review Article – Kavitha Nagandla, Sivalingam Nalliah IeJSME 2014 8(1): 8-18

Insulin requirements in pregnancy and dosage Table 3: Clinical Practice Points In a study by Patterson et al. (2010), the total insulin 1. Rapidly acting analogues, insulin lispro and aspart are safe in requirements through the gestation is noted to vary with pregnancy and improve postprandial glycemic control in women an early rise between 3 and 7 weeks, declining between with pre-gestational diabetes. 7 and 15 weeks and followed by a rise in remainder of 2. When adequate glycemic control is achieved with human insulin pregnancy. This reflects pregnancy related alteration in it is not necessary to convert to rapid acting analogues in view glucose metabolism and nausea and vomiting in early of cost-benefit pregnancy.35 3. Recent studies indicate that long acting analogues has no adverse fetal effects and they effectively reduce the incidence The average insulin requirement in pregnancy is of hypoglycemic episodes. calculated as shown below: 4. Both the drugs (Insulin detemir and glargine) are proved to be safe in pregnancy. However, with lack of definitive fetal Woman’s weight kilograms x k = total insulin beneficial outcome there is no indication for routine use of long requirements: i.e k = units of insulin/kg body weight.36 acting analogues. 5. In women with gestational diabetes and T2DM, with little 1-18 weeks 18-26weeks 26-36weeks 36-40weeks concern for hypoglycemic episodes, there is no evidence to routinely initiate long acting analogues for glycemic control. 0.7unit/kg wt 0.8unit/kg wt 0.9unit/kg wt 1 unit/kg wt

An excess weight gain in pregnancy mandates more Oral hypoglycemic agents insulin and doses of 1.5-2.0units/kg to overcome The pathophysiology of T2DM and gestational the insulin resistance due to pregnancy and obesity. diabetes is due to inadequate insulin secretion and About 2/3 of the total dose is given in the morning (that resistance and therefore the rationale of treating with includes: 33% rapid-acting, 66% intermediate-acting) oral hypoglycemic agents that stimulate the release of and 1/3 in the evening with 50% as rapid-acting insulin insulin from the functional cell of the pancreas, increases before dinner and 50% as intermediate insulin before 37 the insulin sensitivity of peripheral tissues, and as act as bed. insulin secretogogues.38 Although traditionally insulin is Calculating total 24h hours insulin requirement and considered the gold standard in the management due to dose distribution in a pregnant women at 37 weeks its efficacy in achieving euglycemia, it may still prove to gestation with a body weight of 70 kg be inconvenient owing to its cost and invasiveness of the therapy.39 1. First calculate her 24-hour total dose: 0.9 x 70 = ~54 units’ total insulin per day. Evidence for the use of oral hypoglycemic agents in 2. Give 2/3 total dose (36 units) in the AM and 1/3 pregnancy (18 units) in the PM The second generation , glyburide has extensive evidence in pregnancy. In an in vitro study Type of Insulin Pre-Breakfast Pre-Dinner Before Bed with single-cotyledon placental model, glyburide was Rapid acting Insulin- 12 Units 8 units found to only marginally cross the placenta. This was Lispro or Aspart followed by a landmark randomised trial by Langer et al Intermediate acting 24 units 10 units insulin -NPH (2000) which observed that that there was no frequency NPH before bed: Usually requires doses more than 50% to of macrosomia or neonatal hypoglycaemia or maternal prevent dawn or early morning hyperglycemia adverse effects with the use of glyburide in pregnancy.

14 Review Article – Kavitha Nagandla, Sivalingam Nalliah IeJSME 2014 8(1): 8-18

The study demonstrated that there was no detectable was observed that metformin exposed infants had more glyburide level in the umbilical cord despite reaching subcutaneous fat and less visceral fat that may translate therapeutic concentrations of the drug in the maternal to increased insulin sensitivity pattern of growth in blood. However, on close observation it was identified future.45 Regarding and that the study failed to achieve normoglycmia and the in pregnancy ,there is paucity of data with regards to cord samples were inadequate to evaluate its safety.40 their safety.38 Moving forward, few retrospective trials on glyburide Intrapartum management in pregnancy have emerged with the facts that glyburide still fails in 20% of clinical population and The target range of intrapartum glycemic control in the need for subsequent insulin therapy and the rate TIDM, T2DM and gestational diabetes is 3.9-6.1 mmol/l. of neonatal hypoglycaemia and hyperbilirubinemia is This level is determined from the neonatal outcomes increased when compared to insulin.41 Recent studies studies of various observational studies in TIDM women on pharmacokinetics of glyburide, suggest that the and is the recommendation by the American College of peak action is between 2 to 4 hours following the Obstetrician and Gynaecologist.13 administration of the drug. The fact that peak in glucose occurs at 90 minutes after taking a meal suggests that Insulin and glucose requirements during labour glyburide should be administered at least 60 minutes prior Insulin requirements vary in patients with overt 42,43 to a meal to prevent postprandial glucose excursions. DM (T1DM, T2DM) and gestational diabetes. In The questions that remain elusive are whether glyburide women with T1DM there is no endogenous insulin affects the long term wellbeing of the newborn and the whereas women T2DM and gestational diabetes have future progression of gestational diabetes to metabolic sufficient endogenous basal insulin that decreases the syndrome. Until then it is pragmatic to counsel women complications of diabetic ketoacidosis. Thus women on the limited information available albeit the fact with T1DM will require basal insulin infusion to that glyburide is an alternative choice in women with maintain euglycemia during latent phase of spontaneous gestational diabetes, who fail diet therapy and unable to and induced labour.46 comply with insulin therapy. Labour in general has a glucose lowering effect. In the The other oral hypoglycemic agent that is latent phase of labour, the maternal metabolic demands demonstrated to be effective is metformin especially in are not increased and excess glucose is not mandatory. conditions such as polycystic ovarian syndrome, where As active labour ensues, the glucose requirement is it reduces incidence of pregnancy loss and onset of similar to sustained and vigorous exercise with rapid gestational diabetes by improving the insulin sensitivity. depletion of hepatic glycogen stores. It is identified The benefits of using metformin in pregnancy are that there is a need for an eight fold increase in glucose further substantiated in the randomised controlled substrates to meet this demand of glucose which is trial by Rowen et al (2008) which is the ‘Metformin in achieved by glucose infusion at the rate of 2.5mg/kg/mt gestational diabetes trial (MiG trial)’. It was stated that to maintain maternal euglycemia.47 metformin is not associated with increased perinatal complications as compared with insulin and the women What’s new in glucose monitoring intrapartum? preferred metformin to insulin treatment.44 The standard approach is monitoring of capillary In a follow up study of the exposed children at 2 years blood glucose concentration at every two to four hours in in the “ The offspring follow up trial” (TOFU trial) it latent phase and every one to two hours in active phase,

15 Review Article – Kavitha Nagandla, Sivalingam Nalliah IeJSME 2014 8(1): 8-18

and with insulin infusion, it is every hour monitoring. caesarean section. Women admitted for induction The continuous glucose monitoring system (CGMS) is should have their normal insulin doses the night before, a technology that measures the interstitial glucose every once admitted they need short acting insulin to 1 to 10 minutes for a maximum of 3 to 6 days and provides cover the meals and intravenous sliding scale once a comprehensive assessment of glycemic control than labour established. In a comparative study of constant the intermittent monitoring. The reliability, specificity intravenous insulin infusion with continuous and accuracy of CGMS intrapartum was looked at in subcutaneous insulin infusion pump (CSIIP) intrapartum two pilot studies, in which the authors concluded that in 28 women, Feldberg et al (1988) concluded that there were no episodes of neonatal hypoglycemia or CSIIP was superior in achieving and maintaining respiratory distress syndrome and all women reported intrapartum optimal metabolic control, reducing the feeling secure checking their glucose concentrations in incidence of acute fetal distress, caesarean section and real time. CGMS provides information trends rather neonatal hypoglycaemia.51 Epidural analgesia for pain than at point data. There is a delay of about 10 minutes relief in labour has significant advantages as it decreases when compared to capillary blood sugar determination the maternal endogenous catecholamine release and and patient worn devices may create unnecessary anxiety indirectly increases the placental blood flow, reduces the should there be ‘hypo- and hyperglycemia’. The device maternal lactic acid and hence fetal acidosis.52 is expensive and may be inconvenient as it has to be worn for 3-7 days depending on the make of the device. Anaesthesia implications: Perioperative considerations However preliminary studies in pregnancy shows that The consequence of surgery and anaesthesia in a physician monitored CGMS improved hypoglycaemia diabetic patient stimulates a neuroendocrine stress awareness. We need more large prospective trials to mechanism with the production of counter regulatory determine its feasibility and cost-effectiveness before hormones that can result in hyperglycemia and ketosis. incorporating in the protocol of intrapartum glycemic 48, 49 Hence the goal of perioperative management is monitoring. preventing hyperglycemia, hypoglycemia, prevention of ketosis and maintenance of adequate hydration and Insulin and infusion in labour 47 electrolyte balance. For caesarean section, regional Insulin is not required in majority of women with GDM anaesthesia is indicted compared to general and there controlled with diet. In the event of antepartum insulin, is no specific preference for spinal over the epidural only a minority will require insulin in labour. In an anaesthesia. Either spinal or epidural anaesthesia may audit of routine practice of insulin and glucose infusion be suitable in diabetic pregnancies.52 A dextrose free in labour in pre-gestational and gestational diabetes, fluid such as normal saline is used for hydration prior to Barett et al (2009) identified that neonatal induction of anaesthesia as large glucose bolus reduces hypoglycaemia is common with blood sugar levels the umbilical cord pH and neonatal hypoglycaemia. between 4-8 mmol/l. The authors suggested that with Sanjay Datta and his colleagues demonstrated that spinal relaxed capillary blood glucose targets, only 2 % of anaesthesia is a safe technique for caesarean delivery in women on diet and 3.5% of women on insulin in diabetics, with particular caution to avoid hypotension gestational diabetes require insulin therapy in labour. and correction of acute hydration with dextrose free This calls for considering more conservative approach solutions. If the surgery is prolonged, the glucose over the aggressive regimens in these women.50 should be monitored as hyperglycemia during surgery Sliding scale insulin in considered in women with pre- is associated with risk of postoperative wound infection gestational diabetes in labour and delivery by elective and neonatal hypoglycaemia.53 There are additional

16 Review Article – Kavitha Nagandla, Sivalingam Nalliah IeJSME 2014 8(1): 8-18

risks associated with long standing diabetes related to 6. Crowther CA, Hiller JE, Moss JR, et al. Australian Carbohydrate autonomic neuropathy such as orthostatic hypotension, Intolerance Study in Pregnant Women (ACHOIS) Trial Group Effect of treatment of gestational diabetes mellitus on pregnancy painless MI, reduced response to medication-atropine outcomes. N Engl J Med 2005; 352: 2477–86. and propranolol, resting tachycardia, decreased cough 7. Landon MB, Spong CY, Thom E, et al. A multicenter, randomized reflex threshold, increased incidence of obstructive trial of treatment for mild gestational diabetes. N Engl J Med 2009; 361: 1339–48. sleep apnoea and gastroparesis. A thorough preoperative 8. California Diabetes in Pregnancy Project Website. http://www. assessment with history, examination and investigations cdph.ca.gov/programs/cdapp/Pages/default.aspx, including CDPP is paramount to evaluate for end organ damage.52 Fact Sheet http://www.cdph.ca.gov/data/statistics/Documents/ MOCDAPP-GDM-FactSheet.pdf (Accessed July 2010). 9. Castorino1 K, Jovanovic L. Pregnancy and Diabetes Management: Future directions Advances and Controversies. Clin Chem 2011; 57: 2. 10. Hayfaa A Wahabi, Rasmieh A Alzeidan and Samia A Esmaeil. The research in diabetes is rapidly evolving with new Pre-pregnancy care for women with pre-gestational diabetes mellitus: modalities in the diagnosis, monitoring and treatment. a systematic review and meta-analysis. BMC Pub Hlth 2012; 12: 792. There are undoubted benefits of the insulin analogues and 11. American Diabetes Association: Preconception care of women with diabetes. Diabetes Care 2004; 27: 76-8. oral hypoglycaemic agents in pregnancy as evidenced so 12. NICE Clinical Guideline 63. Diabetes in pregnancy: management of far. With the outset of risk involved in clinical trials in diabetes and its complication from pre-conception to the postnatal pregnancy, it can perhaps be unrealistic to expect more period [http://www.nice.org.uk/nicemedia/pdf/CG063Guidance.pdf] 13. ACOG Practice Bulletin. Clinical management guidelines for information on the safety profiles of these drugs. At the obstetrician-gynecologists. Number 30, September 2001 (replaces moment, clinicians have to rely on their knowledge Technical Bulletin Number 200, December 1994). Gestational on the current evidence available on the diagnostic diabetes. Obstet Gynecol 2001; 98: 525-38. 14. Centres for Disease Control and Prevention . National Diabetes approaches and pharmacodynamics and the outcomes of fact sheet: national estimates and general information on diabtes randomised trials that will guide them on the decision and prediabtes in the United States, 2001; Atlanta, GA. USA making process in pregnancy. There are few studies that Department of Health and Human Services, Centres for Disease Control and Prevention 2001. demonstrate that portal insulin delivery and inhaled 15. Banting FG, Best CH, Collip JB, et al. “Pancreatic Extracts in the insulin delivery have dynamics closer to endogenous Treatment of Diabetes Mellitus”. Can Med Assoc J 1922; 12: 141–6. insulin and are expected to make the management of 16. Davis S, Granner D. Insulin, oral hypoglycemic agents, and diabetes much easier. However, we need to synthesise the pharmacology of the endocrine pancreas. In: Hardman JG, Limbird LE, eds. Goodman and Gilman’s: The Pharmacologic Basis of high evidence on safety and efficacy of these routes of Therapeutics. 9th ed. New York, NY: McGraw-Hill; 1996: 1487-518. delivery in pregnant women. 17. FDA/CDER resources page. Food and Drug Administration Web site. Available at: www.fda.gov/cder/drug/shortages/default.htm (Accessed REFERENCES July 14, 2005). 1. Metzger BE, Gabbe SG, Persson B, Buchanan TA, Catalano PA, 18. Menon RK, Cohen RM, Sperling MA, Cutfield WS, Mimouni F, Damm P, Dyer AR, Leiva Ad, Hod M, Kitzmiler JL, Lowe LP, McIntyre Khoury JC. Transplacental passage of insulin in pregnant women with HD, Oats JJ, Omori Y, Schmidt MI. International Association of insulin-dependent diabetes mellitus: Its role in fetal macrosomia. N Diabetes and Pregnancy Study Groups Consensus Panel. Diabetes Engl J Med 1990; 323: 309–15. Care. 2010; 33(3): 676-82. 19. Balsells M, Corcoy R, Mauricio D, Morales J, et al. Insulin antibody 2. American Diabetes Association. Diagnosis an classification of response to a short course of human insulin therapy in women with diabetes mellitus. Diabetic Care 2013; 36 :118-751. gestational diabetes. Diabetes Care 1997; 20: 72-5. 3. American Diabetes Association. Standards of Medical Care in 20. D. R. McCance, P. Damm, E. R. Mathiesen & M. Hod et al. Diabetes—2013. Diabetes Care 2013; 36 Suppl 1: S11-S66. Evaluation of insulin antibodies and placental transfer of insulin 4. Vandorsten JP, Dodson WC, Espeland MA, et al. NIH Consens State aspart in pregnant women with mellitus. Diabetologia Sci Statements. NIH consensus development conference: diagnosing 2008; 51: 2141–3. gestational diabetes mellitus 2013; 29: 1-31. 21. Zinman B. The physiologic replacement of insulin: an elusive goal. 5. Metzger BE, Lowe LP, Dyer AR. Hyperglycemia and adverse N Engl J Med 1989; 321: 363–70. pregnancy outcomes. N Engl J Med 2008; 358: 1991-2002.

17 Review Article – Kavitha Nagandla, Sivalingam Nalliah IeJSME 2014 8(1): 8-18

22. Branko Novak, Ivana Pavlic- Renar, Zeljko Metelko. Treatment of 37. California Diabetes and Pregnancy Program, Sweet Success: diabetes in pregnancy. Diabetologia Croatica 2004; 33:1. Guidelines for Care, 2002. 23. Angelina L. Trujillo. Insulin analogs and preganancy. Diabetes 38. Kavitha, N, Somsubhra De, Kanagasabai S. Oral Hypoglycemic Spectrum 2007; 20: 82-3. Agents in pregnancy: An Update. The Journal of Obstetrics and 24. Anderson JH Jr, Brunelle RL, Koivisto VA, Pfutzner A, et al. Gynecology of India 2013; 63: 82-7. Reduction of postprandial hyperglycemia and frequency of 39. Garcia-Bournissen F, Feig D, Koren G. Maternal-fetal transport of hypoglycemia in IDDM patients on insulin-analog treatment. hypoglycaemic drug. Clin Pharmacokinet. 2003; 42: 303–13. Diabetes 1997; 46: 265–70. 40. Langer, O., Conway, D.L., Berkus, M.D., et al. A Comparison 25. Wyatt J, Frias J, Hoyme H, et al. Congenital anomaly rate in offspring Glyburide and Insulin in Women with Gestational Diabetes Mellitus. of mothers with diabetes treated with insulin lispro during pregnancy. N Engl J Med 2000; 343: 1134-8. Diabet Med 2005; 22: 803–7. 41. Moore TR. Glyburide for the Treatment of Gestational Diabetes. 26. Jovanovic L, Ilic S, Pettitt DJ, Hugo K, et al. The metabolic and Diabetes Care 2007; 30 (Suppl 2): S209-213. immunologic effects of insulin lispro in gestational diabetes. Diabetes 42. Caritis SN, Hebert MF. A pharmacologic approach to the use of Care 1999; 22: 1422–6. glyburide in pregnancy. Obstet Gynecol. 2013; 121: 1309-12. 27. Mathiesen ER, Kinsley B, Amiel SA, et al. Maternal glycemic control 43. Ben-Haroush A, Yogev Y, Chen R, Rosenn B, Hod M, Langer O: The and hypoglycemia in type 1 diabetic pregnancy: a randomized trial of postprandial glucose profile in the diabetic pregnancy. Am J Obstet insulin aspart versus human insulin in 322 pregnant women. Diabetes Gynecol 2004; 191: 576–81. Care 2007; 30: 771–6. 44. Rowan JA, Hague WM, Gao W, Battin MR, Moore MP; MiG 28. Battacharyya A, Brown S, Hughes S, Vice PA. Insulin lispro and Trial Investigators. Metformin versus insulin for the treatment of regular insulin in pregnancy. Q J Med 2001; 94: 255–60. gestational diabetes. N Engl J Med. 2008; 358: 2003-15. 29. V Balaji , V Seshiah. Management of Diabetes in Pregnancy. JAPI 45. Rowan JA, Rush EC, Obolonkin V, et al. Metformin in Gestational 2011; 59. Diabetes: The Offspring Follow-Up (MiG TOFU). Body composition 30. Callesen NF, Damm J, Mathiesen JM, et al. Treatment with the long- at 2 years of age. Diabetes Care 2011; 34 : 2279-84. acting insulin analogues detemir or glargine during pregnancy in 46. Jovanovic L. Glucose and insulin requirements during labor and women with type 1 diabetes: comparison of glycaemic control and delivery: the case for normoglycemia in pregnancies complicated by pregnancy outcome. J Matern Fetal Neonatal Med. 2013; 26: 588-92. diabetes. Endocr Pract 2004; 10: 40. 31. Kurtzhals P, Schaffer L, Sorensen A, et al. Correlation of receptor 47. Kjos SL. Intraparum and postpartum management of insulin and binding and metabolic and mitogenic potencies of insulin analogues blood glucose. www.uptodate.com 2013. designed for clinical use. Diabetes 2000; 49: 999–1005. 48. Byrne EZ, Zisser HC, Jovanovič L. Continuous glucose monitoring: is 32. Lepercq J, Lin J, Hall GC, Wang E, et al. Meta-Analysis of Maternal it helpful in pregnancy? Curr Diabetes Rev. 2008; 4: 223–6. and Neonatal Outcomes Associated with the Use of Insulin Glargine 49. Helen R Murphy. Effectiveness of continuous glucose monitoring in versus NPH Insulin during Pregnancy. Obstet Gynecol Int. 2012; pregnant women with diabetes: randomised clinical trial BMJ 2008; 649070. 337: a1680. 33. Mathiesen ER, Hod M, Ivanisevic M, Duran Garcia S. Maternal 50. Barrett HL, Morris J, McElduff A. Watchful waiting: a management efficacy and safety outcomes in a randomized, controlled trial protocol for maternal glycaemia in the peripartum period. Aust N Z J comparing insulin detemir with NPH insulin in 310 pregnant women Obstet Gynaecol. 2009 ;49:162-7. with type 1 diabetes. Detemir in Pregnancy Study Group. Diabetes 51. Feldberg D, Dicker D, Samuel N, et al . Intrapartum management of Care 2012; 35 : 2012-7. insulin-dependent diabetes mellitus (IDDM) gestants. A comparative 34. Michael F Greene, Rhonda Bentley-Lewis. Medical management study of constant intravenous insulin infusion and continuous of type 1 and mellitus in pregnant women. subcutaneous insulin infusion pump (CSIIP). Acta Obstet Gynecol www.uptodate.com. 2013. Scand. 1988; 67: 333-8. 35. García-Patterson A, Gich I, Amini SB, et al. Insulin requirements 52. Pani N, Mishra SB, and Rath SK. Diabetic parturient – Anaesthetic throughout pregnancy in women with type 1 diabetes mellitus: three implications. Indian J Anaesth. 2010; 54: 387–93. changes of direction. Diabetologia 2010; 53: 446-51. 53. Datta S, Kitzmiller JL, Naulty JS, et al. Acid-Base Status of Diabetic 36. Jovanovic L. Role of diet and insulin treatment of diabetes in Mothers and Their Infants following Spinal Anesthesia for Cesarean pregnancy. Clin Obstet Gynecol 2000; 43: 46-55. Section. A & A 1982; 61: 662-5.

18